NEWS revision 104834
1-*- text -*-
2
3Changes in 2.13:
4
5Support for the Fujitsu FRV architecture added by Red Hat. Models for FR400 and
6FR500 included.
7
8Support for DLX processor added.
9
10GASP has now been deprecated and will be removed in a future release.  Use the
11macro facilities in GAS instead.
12
13GASP now correctly parses floating point numbers. Unless the base is explicitly
14specified, they are interpreted as decimal numbers regardless of the currently
15specified base.
16
17Changes in 2.12:
18
19Support for Don Knuth's MMIX, by Hans-Peter Nilsson.
20
21Support for the OpenRISC 32-bit embedded processor by OpenCores.
22
23The ARM assembler now accepts -march=..., -mcpu=... and -mfpu=... for
24specifying the target instruction set.  The old method of specifying the
25target processor has been deprecated, but is still accepted for
26compatibility.
27
28Support for the VFP floating-point instruction set has been added to
29the ARM assembler.
30
31New psuedo op: .incbin to include a set of binary data at a given point
32in the assembly.  Contributed by Anders Norlander.
33
34The MIPS assembler now accepts -march/-mtune. -mcpu has been deprecated
35but still works for compatability.
36
37The MIPS assembler no longer issues a warning by default when it
38generates a nop instruction from a macro. The new command line option
39-n will turn on the warning.
40
41Changes in 2.11:
42
43x86 gas now supports the full Pentium4 instruction set.
44
45Support for AMD x86-64 architecture, by Jan Hubicka, SuSE Labs.
46
47Support for Motorola 68HC11 and 68HC12.
48
49Support for Texas Instruments TMS320C54x (tic54x).
50
51Support for IA-64.
52
53Support for i860, by Jason Eckhardt.
54
55Support for CRIS (Axis Communications ETRAX series).
56
57x86 gas has a new .arch pseudo op to specify the target CPU architecture.
58
59x86 gas -q command line option quietens warnings about register size changes
60due to suffix, indirect jmp/call without `*', stand-alone prefixes, and
61translating various deprecated floating point instructions.
62
63Changes in 2.10:
64
65Support for the ARM msr instruction was changed to only allow an immediate
66operand when altering the flags field.
67
68Support for ATMEL AVR.
69
70Support for IBM 370 ELF.  Somewhat experimental.
71
72Support for numbers with suffixes.
73
74Added support for breaking to the end of repeat loops.
75
76Added support for parallel instruction syntax (DOUBLEBAR_PARALLEL).
77
78New .elseif pseudo-op added.
79
80New --fatal-warnings option.
81
82picoJava architecture support added.
83
84Motorola MCore 210 processor support added.
85
86A new pseudo-op .intel_syntax has been implemented to allow gas to parse i386
87assembly programs with intel syntax.
88
89New pseudo-ops .func,.endfunc to aid in debugging user-written assembler code.
90
91Added -gdwarf2 option to generate DWARF 2 debugging information.
92
93Full 16-bit mode support for i386.
94
95Greatly improved instruction operand checking for i386.  This change will
96produce errors or warnings on incorrect assembly code that previous versions of
97gas accepted.  If you get unexpected messages from code that worked with older
98versions of gas, please double check the code before reporting a bug.
99
100Weak symbol support added for COFF targets.
101
102Mitsubishi D30V support added.
103
104Texas Instruments c80 (tms320c80) support added.
105
106i960 ELF support added.
107
108ARM ELF support added.
109
110Changes in 2.9:
111
112Texas Instruments c30 (tms320c30) support added.
113
114The assembler now optimizes the exception frame information generated by egcs
115and gcc 2.8.  The new --traditional-format option disables this optimization.
116
117Added --gstabs option to generate stabs debugging information.
118
119The -a option takes a new suboption, m (e.g., -alm) to expand macros in a
120listing.
121
122Added -MD option to print dependencies.
123
124Changes in 2.8:
125
126BeOS support added.
127
128MIPS16 support added.
129
130Motorola ColdFire 5200 support added (configure for m68k and use -m5200).
131
132Alpha/VMS support added.
133
134m68k options --base-size-default-16, --base-size-default-32,
135--disp-size-default-16, and --disp-size-default-32 added.
136
137The alignment directives now take an optional third argument, which is the
138maximum number of bytes to skip.  If doing the alignment would require skipping
139more than the given number of bytes, the alignment is not done at all.
140
141The ELF assembler has a new pseudo-op, .symver, used for symbol versioning.
142
143The -a option takes a new suboption, c (e.g., -alc), to skip false conditionals
144in listings.
145
146Added new pseudo-op, .equiv; it's like .equ, except that it is an error if the
147symbol is already defined.
148
149Changes in 2.7:
150
151The PowerPC assembler now allows the use of symbolic register names (r0, etc.)
152if -mregnames is used.  Symbolic names preceded by a '%' (%r0, etc.) can be
153used any time.  PowerPC 860 move to/from SPR instructions have been added.
154
155Alpha Linux (ELF) support added.
156
157PowerPC ELF support added.
158
159m68k Linux (ELF) support added.
160
161i960 Hx/Jx support added.
162
163i386/PowerPC gnu-win32 support added.
164
165SCO ELF support added.  For OpenServer 5 targets (i386-unknown-sco3.2v5) the
166default is to build COFF-only support.  To get a set of tools that generate ELF
167(they'll understand both COFF and ELF), you must configure with
168target=i386-unknown-sco3.2v5elf.
169
170m88k-motorola-sysv3* support added.
171
172Changes in 2.6:
173
174Gas now directly supports macros, without requiring GASP.
175
176Gas now has an MRI assembler compatibility mode.  Use -M or --mri to select MRI
177mode.  The pseudo-op ``.mri 1'' will switch into the MRI mode until the ``.mri
1780'' is seen; this can be convenient for inline assembler code.
179
180Added --defsym SYM=VALUE option.
181
182Added -mips4 support to MIPS assembler.
183
184Added PIC support to Solaris and SPARC SunOS 4 assembler.
185
186Changes in 2.4:
187
188Converted this directory to use an autoconf-generated configure script.
189
190ARM support, from Richard Earnshaw.
191
192Updated VMS support, from Pat Rankin, including considerably improved debugging
193support.
194
195Support for the control registers in the 68060.
196
197Handles (ignores) a new directive ".this_GCC_requires_the_GNU_assembler", to
198provide for possible future gcc changes, for targets where gas provides some
199features not available in the native assembler.  If the native assembler is
200used, it should become obvious pretty quickly what the problem is.
201
202Usage message is available with "--help".
203
204The GNU Assembler Preprocessor (gasp) is included.  (Actually, it was in 2.3
205also, but didn't get into the NEWS file.)
206
207Weak symbol support for a.out.
208
209A bug in the listing code which could cause an infinite loop has been fixed.
210Bugs in listings when generating a COFF object file have also been fixed.
211
212Initial i386-svr4 PIC implementation from Eric Youngdale, based on code by Paul
213Kranenburg.
214
215Improved Alpha support.  Immediate constants can have a much larger range now.
216Support for the 21164 has been contributed by Digital.
217
218Updated ns32k (pc532-mach, netbsd532) support from Ian Dall.
219
220Changes in 2.3:
221
222Mach i386 support, by David Mackenzie and Ken Raeburn.
223
224RS/6000 and PowerPC support by Ian Taylor.
225
226VMS command scripts (make-gas.com, config-gas.com) have been worked on a bit,
227based on mail received from various people.  The `-h#' option should work again
228too.
229
230HP-PA work, by Jeff Law.  Note, for the PA, gas-2.3 has been designed to work
231with gdb-4.12 and gcc-2.6.  As gcc-2.6 has not been released yet, a special
232version of gcc-2.5.8 has been patched to work with gas-2.3.  You can retrieve
233this special version of gcc-2.5.8 via anonymous ftp from jaguar.cs.utah.edu
234in the "dist" directory.
235
236Vax support in gas fixed for BSD, so it builds and seems to run a couple simple
237tests okay.  I haven't put it through extensive testing.  (GNU make is
238currently required for BSD 4.3 builds.)
239
240Support for the DEC Alpha, running OSF/1 (ECOFF format).  The gas support is
241based on code donated by CMU, which used an a.out-based format.  I'm afraid the
242alpha-a.out support is pretty badly mangled, and much of it removed; making it
243work will require rewriting it as BFD support for the format anyways.
244
245Irix 5 support.
246
247The test suites have been fixed up a bit, so that they should work with a
248couple different versions of expect and dejagnu.
249
250Symbols' values are now handled internally as expressions, permitting more
251flexibility in evaluating them in some cases.  Some details of relocation
252handling have also changed, and simple constant pool management has been added,
253to make the Alpha port easier.
254
255New option "--statistics" for printing out program run times.  This is intended
256to be used with the gcc "-Q" option, which prints out times spent in various
257phases of compilation.  (You should be able to get all of them printed out with
258"gcc -Q -Wa,--statistics", I think.)
259
260----------------------------------------------------------------
261
262Changes in 2.2:
263
264RS/6000 AIX and MIPS SGI Irix 5 support has been added.
265
266Configurations that are still in development (and therefore are convenient to
267have listed in configure.in) still get rejected without a minor change to
268gas/Makefile.in, so people not doing development work shouldn't get the
269impression that support for such configurations is actually believed to be
270reliable.
271
272The program name (usually "as") is printed when a fatal error message is
273displayed.  This should prevent some confusion about the source of occasional
274messages about "internal errors".
275
276ELF support is falling into place.  Support for the 386 should be working.
277Support for SPARC Solaris is in.  HPPA support from Utah is being integrated.
278
279Symbol values are maintained as expressions instead of being immediately boiled
280down to add-symbol, sub-symbol, and constant.  This permits slightly more
281complex calculations involving symbols whose values are not alreadey known.
282
283DBX-style debugging info ("stabs") is now supported for COFF formats.
284If any stabs directives are seen in the source, GAS will create two new
285sections: a ".stab" and a ".stabstr" section.  The format of the .stab
286section is nearly identical to the a.out symbol format, and .stabstr is
287its string table.  For this to be useful, you must have configured GCC
288to generate stabs (by defining DBX_DEBUGGING_INFO), and must have a GDB
289that can use the stab sections (4.11 or later).
290
291LynxOS, on i386 and m68k platforms, is now supported.  SPARC LynxOS
292support is in progress.
293
294----------------------------------------------------------------
295
296Changes in 2.1:
297
298Several small fixes for i386-aix (PS/2) support from Minh Tran-Le have been
299incorporated, but not well tested yet.
300
301Altered the opcode table split for m68k; it should require less VM to compile
302with gcc now.
303
304Some minor adjustments to add (Convergent Technologies') Miniframe support,
305suggested by Ronald Cole.
306
307HPPA support (running OSF only, not HPUX) has been contributed by Utah.  This
308includes improved ELF support, which I've started adapting for SPARC Solaris
3092.x.  Integration isn't completely, so it probably won't work.
310
311HP9000/300 support, donated by HP, has been merged in.
312
313Ian Taylor has finished the MIPS ECOFF (Ultrix, Irix) support.
314
315Better error messages for unsupported configurations (e.g., hppa-hpux).
316
317Test suite framework is starting to become reasonable.
318
319----------------------------------------------------------------
320
321Changes in 2.0:
322
323Mostly bug fixes.
324
325Some more merging of BFD and ELF code, but ELF still doesn't work.
326
327----------------------------------------------------------------
328
329Changes in 1.94:
330
331BFD merge is partly done.  Adventurous souls may try giving configure the
332"--with-bfd-assembler" option.  Currently, ELF format requires it, a.out format
333accepts it; SPARC CPU accepts it.  It's the default only for OS "elf" or
334"solaris".  (ELF isn't really supported yet.  It needs work.  I've got some
335code from Utah for HP-PA ELF, and from DG for m88k ELF, but they're not fully
336merged yet.)
337
338The 68K opcode table has been split in half.  It should now compile under gcc
339without consuming ridiculous amounts of memory.
340
341A couple data structures have been reduced in size.  This should result in
342saving a little bit of space at runtime.
343
344Support for MIPS, from OSF and Ralph Campbell, has been merged in.  The OSF
345code provided ROSE format support, which I haven't merged in yet.  (I can make
346it available, if anyone wants to try it out.)  Ralph's code, for BSD 4.4,
347supports a.out format.  We don't have ECOFF support in just yet; it's coming.
348
349Support for the Hitachi H8/500 has been added.
350
351VMS host and target support should be working now, thanks chiefly to Eric
352Youngdale.
353
354----------------------------------------------------------------
355
356Changes in 1.93.01:
357
358For m68k, support for more processors has been added: 68040, CPU32, 68851.
359
360For i386, .align is now power-of-two; was number-of-bytes.
361
362For m68k, "%" is now accepted before register names.  For COFF format, which
363doesn't use underscore prefixes for C labels, it is required, so variable "a0"
364can be distinguished from the register.
365
366Last public release was 1.38.  Lots of configuration changes since then, lots
367of new CPUs and formats, lots of bugs fixed.
368
369
370Local variables:
371fill-column: 79
372End:
373